This is basically an introductory course on AI as told through the history of Disney's animatronics. There is a companion video which covers the early decades of robotics and focuses on Walt's futurism; this longer video focuses on how AI has attempted to pull money out of customer wallets delight park visitors by putting smiles onto faces. Perjurer focuses on concrete examples; there's no talk of hyperreality here, although there is a bit of theory-building which fits each example into a generic framework for understanding conversations.

The video has too many good sneers for me to choose. A common theme is guests tricking AI hosts into behaving inappropriately. There's this theme of the robots only functioning properly within controlled conditions, as if every robot were its own science experiment. This lines up with what I've seen in manufacturing and logistics; robots sure can work fast but they are inflexible, pre-programmed, and highly sensitive to unexpected variance in their environment.

No, I take it back. Listening to E.T. say "D-D-D-D-D-D-" or "lasagna, lasagna" is very funny. Skip to the interlude about Universal Studios for that.

Of course, little of this is truly new, but it's nice to see a version of this history which puts everything together to point out that Disney's goal of creating robots which imitate inhuman characters is fucked-up horror but which isn't a fucked-up horror story. Going in the other direction, an AI-skeptical viewpoint could maybe make those stories more interesting.